Medicine
®
OBSERVATIONAL STUDY
Influence of Preoperative Chemoradiotherapy on the Surgical Strategy According to the Clinical T Stage of Patients With Rectal Cancer In Ja Park, MD, PhD, Jong Lyul Lee, MD, Yong Sik Yoon, MD, PhD, Chan Wook Kim, MD, Seok-Byung Lim, MD, PhD, Jong Seok Lee, MD, PhD, Seong Ho Park, MD, PhD, Jin Hong Park, MD, PhD, Jong Hoon Kim, MD, PhD, Chang Sik Yu, MD, PhD, and Jin Cheon Kim, MD, PhD
Abstract: The aim of this study was to evaluate the pathologic responses and changes to surgical strategies following preoperative chemoradiotherapy (PCRT) in rectal cancer patients according to their clinical T stage (cT). The use of PCRT has recently been extended to less advanced disease. The authors enrolled 650 patients with cT2 to 4 mid and low rectal cancer who received both PCRT and surgical resection. The rate of total regression and the proportion of local excision were compared according to the cT category. The 3-year recurrence-free survival (RFS) rate was compared using the log-rank test according to patient cT category, pathologic stage, and type of surgical treatment. Patients with cT2 were older (P ¼ 0.001), predominately female (P ¼ 0.028), and had low-lying rectal cancer (P ¼ 0.008). Pathologic total regression was achieved most frequently in cT2 patients (54% of cT2 versus 17.6% of cT3 versus 8.2% of cT4; P < 0.001). Local excision was performed on 42 cT2 (42%) and 24 cT3 (5.2%) patients (P < 0.001). The 3-year RFS rates differed according to both cT (P < 0.001) and ypT stage (P < 0.001). Among patients with ypT0 to 1 disease, the 3-year RFS did not differ according to the type of surgical treatment received (P ¼ 0.5). Total regression of the primary tumor and a change in the surgical strategy after PCRT are most commonly seen in cT2 disease. Although PCRT is not generally indicated for cT2 rectal cancer, optimal surgical treatment may be achieved with the tailored use of PCRT. (Medicine 94(52):e2377) Abbreviations: AJCC = American Joint Committee on Cancer, CT = computed tomography, FU = fluorouracil, IQR = interqurtile
Editor: Samantha Martin. Received: August 18, 2015; revised: November 23, 2015; accepted: December 4, 2015. From the Department of Colon and Rectal Surgery (IJP, JLL, YSY, CWK, S-BL, JCK); Department of Radiology (JSL, SHP); and Department of Radiation Oncology (JHP, JHK), University of Ulsan College of Medicine and Asan Medical Center, Seoul, South Korea. Correspondence: In Ja Park, MD, PhD, Department of Colon and Rectal Surgery, University of College of Medicine and Asan Medical Center, 86 Asanbyeongwon-gil, Songpa-gu, Seoul 138-736, South Korea (e-mail: [email protected]). This study was supported by a grant (2014–240) from the Asan Institute for Life Sciences, Seoul, South Korea. The authors have no conflicts of interest to disclose. Copyright # 2015 Wolters Kluwer Health, Inc. All rights reserved. This is an open access article distributed under the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. ISSN: 0025-7974 DOI: 10.1097/MD.0000000000002377
Medicine
Volume 94, Number 52, December 2015
range, LE = local excision, LNs = lymph nodes, MRI = magnetic resonance imaging, PCRT = preoperative chemoradiotherapy, RFS = recurrence-free survival, TME = total mesorectal excision, TR = total regression, TRG = tumor regression grade.
INTRODUCTION
T
he current standard treatment for locally advanced rectal cancer is preoperative chemoradiotherapy (PCRT), which has been shown to successfully downstage tumors and promote favorable clinical outcomes.1,2 The benefits of PCRT are optimal in patients with tumors that are highly responsive to this treatment and in patients who have certain clinical characteristics that are associated with superior treatment outcomes.3 – 5 Generally, PCRT has been indicated only for advanced rectal cancer [clinically diagnosed as T3 to 4 (cT3 to 4)] and rectal cancer presenting with metastatic lymph nodes (LNs). Recent studies, however, have reported treatment of cT2 rectal cancer using PCRT.6 – 8 The standard of treatment for most stage I rectal cancers is surgery alone, specifically total mesorectal excision (TME).9 Local excision (LE), including transanal excision and transanal endoscopic microsurgery, has been explored as a surgical treatment for stage I disease because of the morbidity and/or functional derangement associated with TME. Local excision alone, however, demonstrates inferior oncologic outcomes in comparison with TME.10,11 Despite continuing reports on inferior outcomes following LE, the rate of LE use to treat stage I tumors has steadily increased.12 The continued use of LE to treat stage I tumors may be because of the lower morbidity rates and better long-term functional outcomes associated with LE compared with TME. For distal rectal cancers in particular, LE offers the promise of sphincter preservation, whereas TME often results in permanent ostomy creation. The use of PCRT in cT2 disease is expected to improve oncologic outcomes comparison with LE, therefore extending the indications of LE to cT2 disease for functional improvement. The influence of surgical strategies might be closely associated with the rate of total regression (TR) of primary rectal tumors because the tumor response to CRT has emerged as an important predictor of tumor control and patient survival.13,14 We evaluated the rate of TR for primary tumors following PCRT and the influence of PCRT on surgical strategies (ie, the rate of LE according to cT category) in patients with mid to low rectal cancer. In addition, we evaluated oncologic outcomes in a series of cT2 rectal patients according to surgical treatment (LE versus TME). www.md-journal.com |
1
Medicine
Park et al
METHODS Patients, Diagnosis, and Clinical Staging We included 650 patients with primary mid to low rectal cancer (located within 10 cm of the anal verge) that received treatment with PCRT followed by surgical resection (including LE) between January 2011 and December 2013 at Asan Medical Center, Seoul, South Korea. Patients with simultaneous distant metastases on pretreatment work-ups or with a prior or concurrent malignancy were excluded. Patients who did not receive any kind of surgical treatment or who were diagnosed using techniques other than magnetic resonance imaging (MRI) were also excluded. The clinical stage of each tumor was diagnosed on MRI using a high spatial resolution phased-array MR technique. An MRI diagnosis of a cT3 lesion was based on the presence of tumor signal intensity that extended through the muscle layers into the perirectal fat with a broad-based bulging configuration and continuity with the intramural portion of the tumor. Tumors located within the muscle layer were diagnosed as cT3 lesions. Tumor signal intensity that extended beyond the perirectal fat or demonstrated a loss of plane between the adjacent organs was diagnosed as a cT4 lesion. Metastatic status of the LNs was ascertained by considering nodal size and morphologic characteristics, such as signal intensity, border, contour, shape, and texture. This practice is in contrast to the method agreed on and practiced by many experts in which a single criterion, such as a size threshold, is evaluated.15
Preoperative Chemoradiotherapy, Surgical Treatment, and Pathologic Examination The PCRT regimen consisted of a 45-Gy dose of pelvic external beam radiation delivered in 25 fractions during 5 weeks. During the last week of treatment, patients received a 5.4-Gy boost to the primary tumor delivered in 5 (second daily) fractions, cumulating in a total radiation dose of 50.4 Gy. Chemotherapy was delivered as 2 cycles via an intravenous bolus of 5-fluorouracil (FU) (375 mg/m2/d) and leucovorin (20 mg/m2/d) for 3 days during the first and fifth weeks of radiation therapy or as oral capecitabine (1650 mg/m2/d), administered twice-daily during radiation therapy. Approximately 4 weeks after completing PCRT, clinical stage was reevaluated using pelvic MRI, abdominopelvic computed tomography, and sigmoidoscopy. Surgical resection was planned within 6 to 8 weeks of PCRT completion. Patients were supposed to undergo radical resection according to the principles of tumor-specific mesorectal excision. Patient refusal of radical surgery and poor performance status were reasons for undergoing LE instead of TME following PCRT. Patients who chose LE were fully informed about the tumor response to PCRT and the surgical options between radical resection and LE. Each patient provided written informed consent before treatment. Adjuvant chemotherapy was recommended for all medically fit patients who received PCRT and radical resection and consisted of infused 5–FU or capecitabine for 6 months. Oxaliplatin-based chemotherapy was administered to some patients based on their postoperative pathologic results.
Pathologic Examination, Follow-up, and Oncologic Outcomes Dedicated gastrointestinal cancer pathologists performed standard pathologic tumor staging. Tumors were pathologically staged according to the guidelines of the AJCC (7th edition). The LNs were identified by manual dissection of mesorectum
2
| www.md-journal.com
Volume 94, Number 52, December 2015
and examined using 1 to 3 separate sections per node. Pathologic responses to PCRT were evaluated in the resected specimens using the tumor regression grade system suggested by the Gastrointestinal Pathology Study Group of the Korean Society of Pathologists.16 Tumor regression was scored as follows; TR with no residual tumor cells and only fibrotic mass, near-total regression with microscopic residual tumor in the fibrotic tissue, and moderate regression with easy-to-find irradiation-related changes with residual tumor; minimal regression with a dominant tumor mass with obvious irradiation-related changes, or no regression or evidence of irradiation-related changes, such as fibrosis, necrosis, or vascular changes. Postoperative follow-up consisted of routine physical examinations and carcinoembryonic antigen measurements every 3 to 6 months, along with abdominal pelvis and chest computed tomography every 6 months to 1 year. Colonoscopies were performed at 6 months or 1-year postoperatively and every 2 to 3 years thereafter. Recurrence-free survival (RFS) was defined as the time between surgery and the first recurrence event or death.
Statistical Analysis Pearson x2 test, Fisher exact test, or Student t test were used for comparison of clinicopathologic characteristic of the patients according to their cT category as applicable. The associations between surgical treatment and pathologic results were also compared between patient groups. Cases with disease recurrence or death from any cause were identified as failures at the time of recurrence or death for RFS analysis. Noncancer deaths were not censored. The 3-year RFS rates were determined using the Kaplan–Meier method, and compared using the log-rank test between groups. Cox proportional hazards regression analysis was used to perform the multivariate comparisons. In all analyses, P < 0.05 was considered statistically significant. All statistical analyses were performed using SPSS version 21.0 (IBM Statistics, Armonk, NY).
RESULTS Patient Characteristics We included 650 patients who met inclusion criteria. The median age was 61 years [interquartile range (IQR) ¼ 48–66 years]. Men (64.6%) were predominant among patients. The median distance of the tumor from the anal verge was 6 cm (IQR ¼ 4–8 cm). Most tumors were cT3 on preoperative staging. Concurrent chemotherapy using of 5-FU were used in 42.4% and capecitabine in 57.6% of the patients. Sixty-seven patients underwent LE, and the remainder underwent TME or tumor-specific mesorectal excision depending on extent and location of the tumor. Sphincter-preserving operations were performed on 81.5% of the patients treated with radical resection. In total, 143 patients (22%) demonstrated TR on tumor regression grade.
Clinicopathologic Characteristics and Surgical Treatment According to Clinical T Stage Category The cT2 group was older (P ¼ 0.001), predominately female (P ¼ 0.028), and demonstrated low-lying rectal cancer (P ¼ 0.008). Patients with a lower cT category were also more likely to demonstrate TR of the primary tumor following PCRT (P < 0.001). Our analyses showed that 54% of patients with cT2 disease demonstrated TR but only 8.2% of patients with cT4 Copyright
#
2015 Wolters Kluwer Health, Inc. All rights reserved.
Medicine
Volume 94, Number 52, December 2015
disease achieved TR of the primary tumor. The pathologic T stage varied according to cT category. Although tumors in patients in the cT2 group demonstrated downstaging to a ypT0 to 1 primary tumor in 79 cases (79%), tumors in patients in the cT3 group were downstaged to a ypT0 to 2 primary tumor in 250 cases (52.8%). A total of 73% of patients with cT4 disease, however, maintained a ypT3 to 4 disease status. Among the patients who underwent radical resection, the ypNþ rate also differed according to cT category. Among patients with cT2 disease, only 1 patient demonstrated a ypNþ tumor (1.7%). The number of ypNþ tumors increased according to the ypT stage. ypNþ tumors accounted for 7 of the 144 ypT0 to 1 tumors (4.9%) that underwent radical resection. Among patients with ypT2 disease, the ypNþ rate was 12.4% but abruptly increased to 39% and 70% in patients with ypT3 and ypT4 disease, respectively. Local excision was more frequently performed on cT2 cases (42%). Sphincter preservation, including LE, was most commonly performed for cT2 disease, although the cT2 group demonstrated more low-lying rectal cancer cases. Among patients with cT3 disease, only 5.2% of cases underwent LE, although 17.6% of these patients demonstrated TR of the primary tumor. No patient with cT4 disease underwent LE. The surgical strategy was changed from the current standard surgical treatment to an alternative method in 77.8% of cT2 and 29.3% of cT3 patients with TR (Table 1).
Accuracy of Pelvic Magnetic Resonance Imaging for Predicting the ypT Stage Following Preoperative Chemoradiotherapy According to the Clinical T Stage Category For 323 patients, pelvic MRI following PCRT accurately predicted the ypT stage. Underestimation of ypT stage occurred
Surgical Strategy After Preoperative Chemoradiotherapy
in 8.2% of patients. The accuracy of post-PCRT MRI, in terms of predicting TR of the primary tumor, differed according to cT category. Among patients with cT2 disease, TR of the primary tumor was predicted in 39.8% of patients, but it was only predicted in 10% of patients with either cT3 or cT4 disease. The accuracy of post-PCRT MRI also differed according to ypT stage. Magnetic resonance imaging could not accurately predict ypT stage in approximately 40% of patients with ypT0 to 2 disease, but this inaccuracy increased to 77.8% among patients with ypT3 disease (Table 2).
Recurrence and Survival The median follow-up period was 30 months (IQR ¼ 21– 39 months) for the entire study cohort and did not differ according to cT category. Overall, recurrence was observed in 115 patients (17.7%). Nine patients demonstrated only local recurrence, 98 patients demonstrated only systemic recurrence, and 8 patients demonstrated both local and systemic recurrence. The lung was the most common initial metastatic site (68 of 115 patients; 59.1%). For the entire cohort, the RFS at 3 years was 79.6%. The 3-year RFS differed according to cT category and ypT stage (Fig. 1). Among patients with ypT0 to 1 disease, the 3-year RFS did not differ according to the type of surgical resection or LN metastasis (Fig. 2). According to the adjusted multivariate Cox regression analysis, the type of surgical resection, sex, age, location, and cT category were not associated with RFS among patients with ypT0 to 1 disease (Table 3).
DISCUSSION Our current study findings indicate that PCRT can result in significantly higher TR of the primary tumor in rectal cancer patients and provide a greater influence on subsequent surgical
TABLE 1. Clinicopathological Characteristics of the Study Patients According to Their Clinical T Stage Category cT2 (n ¼ 100)
Variable Sex Male Female Age, y Tumor location (cm from AV) 5 >5 and 10 ycNþ Tumor regression grade TR No TR ypT ypT0–Tis ypT1 ypT2 ypT3 ypT4 ypNþ Surgery Local excision Radical resection
cT3 (n ¼ 465)
cT4 (n ¼ 85)
P 0.028
53 47 63
(53) (47) (38–79)
312 153 60
(67.1) (32.9) (35–78)
55 30 56
(64.7) (35.3) (32–69)
80 20 42
(80) (20) (42)
299 166 443
(64.3) (35.7) (95.3)
60 25 83
(70.6) (29.4) (97.6)
54 46
(54) (46)
82 383
(17.6) (82.4)
7 78
(8.2) (91.8)
60 19 21 – – 1
(60) (19) (21) – – (1.7)
87 18 145 217 1 113
(18.7) (3.9) (30.5) (46.7) (0.2) (24.3)
10 3 10 53 9 25
(11.9) (3.5) (11.8) (62.4) (10.6) (29.4)
42 58
(42) (58)
24 441
(5.2) (94.8)
0 85
(0) (100)
0.001 0.008
®
OBSERVATIONAL STUDY
Influence of Preoperative Chemoradiotherapy on the Surgical Strategy According to the Clinical T Stage of Patients With Rectal Cancer In Ja Park, MD, PhD, Jong Lyul Lee, MD, Yong Sik Yoon, MD, PhD, Chan Wook Kim, MD, Seok-Byung Lim, MD, PhD, Jong Seok Lee, MD, PhD, Seong Ho Park, MD, PhD, Jin Hong Park, MD, PhD, Jong Hoon Kim, MD, PhD, Chang Sik Yu, MD, PhD, and Jin Cheon Kim, MD, PhD
Abstract: The aim of this study was to evaluate the pathologic responses and changes to surgical strategies following preoperative chemoradiotherapy (PCRT) in rectal cancer patients according to their clinical T stage (cT). The use of PCRT has recently been extended to less advanced disease. The authors enrolled 650 patients with cT2 to 4 mid and low rectal cancer who received both PCRT and surgical resection. The rate of total regression and the proportion of local excision were compared according to the cT category. The 3-year recurrence-free survival (RFS) rate was compared using the log-rank test according to patient cT category, pathologic stage, and type of surgical treatment. Patients with cT2 were older (P ¼ 0.001), predominately female (P ¼ 0.028), and had low-lying rectal cancer (P ¼ 0.008). Pathologic total regression was achieved most frequently in cT2 patients (54% of cT2 versus 17.6% of cT3 versus 8.2% of cT4; P < 0.001). Local excision was performed on 42 cT2 (42%) and 24 cT3 (5.2%) patients (P < 0.001). The 3-year RFS rates differed according to both cT (P < 0.001) and ypT stage (P < 0.001). Among patients with ypT0 to 1 disease, the 3-year RFS did not differ according to the type of surgical treatment received (P ¼ 0.5). Total regression of the primary tumor and a change in the surgical strategy after PCRT are most commonly seen in cT2 disease. Although PCRT is not generally indicated for cT2 rectal cancer, optimal surgical treatment may be achieved with the tailored use of PCRT. (Medicine 94(52):e2377) Abbreviations: AJCC = American Joint Committee on Cancer, CT = computed tomography, FU = fluorouracil, IQR = interqurtile
Editor: Samantha Martin. Received: August 18, 2015; revised: November 23, 2015; accepted: December 4, 2015. From the Department of Colon and Rectal Surgery (IJP, JLL, YSY, CWK, S-BL, JCK); Department of Radiology (JSL, SHP); and Department of Radiation Oncology (JHP, JHK), University of Ulsan College of Medicine and Asan Medical Center, Seoul, South Korea. Correspondence: In Ja Park, MD, PhD, Department of Colon and Rectal Surgery, University of College of Medicine and Asan Medical Center, 86 Asanbyeongwon-gil, Songpa-gu, Seoul 138-736, South Korea (e-mail: [email protected]). This study was supported by a grant (2014–240) from the Asan Institute for Life Sciences, Seoul, South Korea. The authors have no conflicts of interest to disclose. Copyright # 2015 Wolters Kluwer Health, Inc. All rights reserved. This is an open access article distributed under the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. ISSN: 0025-7974 DOI: 10.1097/MD.0000000000002377
Medicine
Volume 94, Number 52, December 2015
range, LE = local excision, LNs = lymph nodes, MRI = magnetic resonance imaging, PCRT = preoperative chemoradiotherapy, RFS = recurrence-free survival, TME = total mesorectal excision, TR = total regression, TRG = tumor regression grade.
INTRODUCTION
T
he current standard treatment for locally advanced rectal cancer is preoperative chemoradiotherapy (PCRT), which has been shown to successfully downstage tumors and promote favorable clinical outcomes.1,2 The benefits of PCRT are optimal in patients with tumors that are highly responsive to this treatment and in patients who have certain clinical characteristics that are associated with superior treatment outcomes.3 – 5 Generally, PCRT has been indicated only for advanced rectal cancer [clinically diagnosed as T3 to 4 (cT3 to 4)] and rectal cancer presenting with metastatic lymph nodes (LNs). Recent studies, however, have reported treatment of cT2 rectal cancer using PCRT.6 – 8 The standard of treatment for most stage I rectal cancers is surgery alone, specifically total mesorectal excision (TME).9 Local excision (LE), including transanal excision and transanal endoscopic microsurgery, has been explored as a surgical treatment for stage I disease because of the morbidity and/or functional derangement associated with TME. Local excision alone, however, demonstrates inferior oncologic outcomes in comparison with TME.10,11 Despite continuing reports on inferior outcomes following LE, the rate of LE use to treat stage I tumors has steadily increased.12 The continued use of LE to treat stage I tumors may be because of the lower morbidity rates and better long-term functional outcomes associated with LE compared with TME. For distal rectal cancers in particular, LE offers the promise of sphincter preservation, whereas TME often results in permanent ostomy creation. The use of PCRT in cT2 disease is expected to improve oncologic outcomes comparison with LE, therefore extending the indications of LE to cT2 disease for functional improvement. The influence of surgical strategies might be closely associated with the rate of total regression (TR) of primary rectal tumors because the tumor response to CRT has emerged as an important predictor of tumor control and patient survival.13,14 We evaluated the rate of TR for primary tumors following PCRT and the influence of PCRT on surgical strategies (ie, the rate of LE according to cT category) in patients with mid to low rectal cancer. In addition, we evaluated oncologic outcomes in a series of cT2 rectal patients according to surgical treatment (LE versus TME). www.md-journal.com |
1
Medicine
Park et al
METHODS Patients, Diagnosis, and Clinical Staging We included 650 patients with primary mid to low rectal cancer (located within 10 cm of the anal verge) that received treatment with PCRT followed by surgical resection (including LE) between January 2011 and December 2013 at Asan Medical Center, Seoul, South Korea. Patients with simultaneous distant metastases on pretreatment work-ups or with a prior or concurrent malignancy were excluded. Patients who did not receive any kind of surgical treatment or who were diagnosed using techniques other than magnetic resonance imaging (MRI) were also excluded. The clinical stage of each tumor was diagnosed on MRI using a high spatial resolution phased-array MR technique. An MRI diagnosis of a cT3 lesion was based on the presence of tumor signal intensity that extended through the muscle layers into the perirectal fat with a broad-based bulging configuration and continuity with the intramural portion of the tumor. Tumors located within the muscle layer were diagnosed as cT3 lesions. Tumor signal intensity that extended beyond the perirectal fat or demonstrated a loss of plane between the adjacent organs was diagnosed as a cT4 lesion. Metastatic status of the LNs was ascertained by considering nodal size and morphologic characteristics, such as signal intensity, border, contour, shape, and texture. This practice is in contrast to the method agreed on and practiced by many experts in which a single criterion, such as a size threshold, is evaluated.15
Preoperative Chemoradiotherapy, Surgical Treatment, and Pathologic Examination The PCRT regimen consisted of a 45-Gy dose of pelvic external beam radiation delivered in 25 fractions during 5 weeks. During the last week of treatment, patients received a 5.4-Gy boost to the primary tumor delivered in 5 (second daily) fractions, cumulating in a total radiation dose of 50.4 Gy. Chemotherapy was delivered as 2 cycles via an intravenous bolus of 5-fluorouracil (FU) (375 mg/m2/d) and leucovorin (20 mg/m2/d) for 3 days during the first and fifth weeks of radiation therapy or as oral capecitabine (1650 mg/m2/d), administered twice-daily during radiation therapy. Approximately 4 weeks after completing PCRT, clinical stage was reevaluated using pelvic MRI, abdominopelvic computed tomography, and sigmoidoscopy. Surgical resection was planned within 6 to 8 weeks of PCRT completion. Patients were supposed to undergo radical resection according to the principles of tumor-specific mesorectal excision. Patient refusal of radical surgery and poor performance status were reasons for undergoing LE instead of TME following PCRT. Patients who chose LE were fully informed about the tumor response to PCRT and the surgical options between radical resection and LE. Each patient provided written informed consent before treatment. Adjuvant chemotherapy was recommended for all medically fit patients who received PCRT and radical resection and consisted of infused 5–FU or capecitabine for 6 months. Oxaliplatin-based chemotherapy was administered to some patients based on their postoperative pathologic results.
Pathologic Examination, Follow-up, and Oncologic Outcomes Dedicated gastrointestinal cancer pathologists performed standard pathologic tumor staging. Tumors were pathologically staged according to the guidelines of the AJCC (7th edition). The LNs were identified by manual dissection of mesorectum
2
| www.md-journal.com
Volume 94, Number 52, December 2015
and examined using 1 to 3 separate sections per node. Pathologic responses to PCRT were evaluated in the resected specimens using the tumor regression grade system suggested by the Gastrointestinal Pathology Study Group of the Korean Society of Pathologists.16 Tumor regression was scored as follows; TR with no residual tumor cells and only fibrotic mass, near-total regression with microscopic residual tumor in the fibrotic tissue, and moderate regression with easy-to-find irradiation-related changes with residual tumor; minimal regression with a dominant tumor mass with obvious irradiation-related changes, or no regression or evidence of irradiation-related changes, such as fibrosis, necrosis, or vascular changes. Postoperative follow-up consisted of routine physical examinations and carcinoembryonic antigen measurements every 3 to 6 months, along with abdominal pelvis and chest computed tomography every 6 months to 1 year. Colonoscopies were performed at 6 months or 1-year postoperatively and every 2 to 3 years thereafter. Recurrence-free survival (RFS) was defined as the time between surgery and the first recurrence event or death.
Statistical Analysis Pearson x2 test, Fisher exact test, or Student t test were used for comparison of clinicopathologic characteristic of the patients according to their cT category as applicable. The associations between surgical treatment and pathologic results were also compared between patient groups. Cases with disease recurrence or death from any cause were identified as failures at the time of recurrence or death for RFS analysis. Noncancer deaths were not censored. The 3-year RFS rates were determined using the Kaplan–Meier method, and compared using the log-rank test between groups. Cox proportional hazards regression analysis was used to perform the multivariate comparisons. In all analyses, P < 0.05 was considered statistically significant. All statistical analyses were performed using SPSS version 21.0 (IBM Statistics, Armonk, NY).
RESULTS Patient Characteristics We included 650 patients who met inclusion criteria. The median age was 61 years [interquartile range (IQR) ¼ 48–66 years]. Men (64.6%) were predominant among patients. The median distance of the tumor from the anal verge was 6 cm (IQR ¼ 4–8 cm). Most tumors were cT3 on preoperative staging. Concurrent chemotherapy using of 5-FU were used in 42.4% and capecitabine in 57.6% of the patients. Sixty-seven patients underwent LE, and the remainder underwent TME or tumor-specific mesorectal excision depending on extent and location of the tumor. Sphincter-preserving operations were performed on 81.5% of the patients treated with radical resection. In total, 143 patients (22%) demonstrated TR on tumor regression grade.
Clinicopathologic Characteristics and Surgical Treatment According to Clinical T Stage Category The cT2 group was older (P ¼ 0.001), predominately female (P ¼ 0.028), and demonstrated low-lying rectal cancer (P ¼ 0.008). Patients with a lower cT category were also more likely to demonstrate TR of the primary tumor following PCRT (P < 0.001). Our analyses showed that 54% of patients with cT2 disease demonstrated TR but only 8.2% of patients with cT4 Copyright
#
2015 Wolters Kluwer Health, Inc. All rights reserved.
Medicine
Volume 94, Number 52, December 2015
disease achieved TR of the primary tumor. The pathologic T stage varied according to cT category. Although tumors in patients in the cT2 group demonstrated downstaging to a ypT0 to 1 primary tumor in 79 cases (79%), tumors in patients in the cT3 group were downstaged to a ypT0 to 2 primary tumor in 250 cases (52.8%). A total of 73% of patients with cT4 disease, however, maintained a ypT3 to 4 disease status. Among the patients who underwent radical resection, the ypNþ rate also differed according to cT category. Among patients with cT2 disease, only 1 patient demonstrated a ypNþ tumor (1.7%). The number of ypNþ tumors increased according to the ypT stage. ypNþ tumors accounted for 7 of the 144 ypT0 to 1 tumors (4.9%) that underwent radical resection. Among patients with ypT2 disease, the ypNþ rate was 12.4% but abruptly increased to 39% and 70% in patients with ypT3 and ypT4 disease, respectively. Local excision was more frequently performed on cT2 cases (42%). Sphincter preservation, including LE, was most commonly performed for cT2 disease, although the cT2 group demonstrated more low-lying rectal cancer cases. Among patients with cT3 disease, only 5.2% of cases underwent LE, although 17.6% of these patients demonstrated TR of the primary tumor. No patient with cT4 disease underwent LE. The surgical strategy was changed from the current standard surgical treatment to an alternative method in 77.8% of cT2 and 29.3% of cT3 patients with TR (Table 1).
Accuracy of Pelvic Magnetic Resonance Imaging for Predicting the ypT Stage Following Preoperative Chemoradiotherapy According to the Clinical T Stage Category For 323 patients, pelvic MRI following PCRT accurately predicted the ypT stage. Underestimation of ypT stage occurred
Surgical Strategy After Preoperative Chemoradiotherapy
in 8.2% of patients. The accuracy of post-PCRT MRI, in terms of predicting TR of the primary tumor, differed according to cT category. Among patients with cT2 disease, TR of the primary tumor was predicted in 39.8% of patients, but it was only predicted in 10% of patients with either cT3 or cT4 disease. The accuracy of post-PCRT MRI also differed according to ypT stage. Magnetic resonance imaging could not accurately predict ypT stage in approximately 40% of patients with ypT0 to 2 disease, but this inaccuracy increased to 77.8% among patients with ypT3 disease (Table 2).
Recurrence and Survival The median follow-up period was 30 months (IQR ¼ 21– 39 months) for the entire study cohort and did not differ according to cT category. Overall, recurrence was observed in 115 patients (17.7%). Nine patients demonstrated only local recurrence, 98 patients demonstrated only systemic recurrence, and 8 patients demonstrated both local and systemic recurrence. The lung was the most common initial metastatic site (68 of 115 patients; 59.1%). For the entire cohort, the RFS at 3 years was 79.6%. The 3-year RFS differed according to cT category and ypT stage (Fig. 1). Among patients with ypT0 to 1 disease, the 3-year RFS did not differ according to the type of surgical resection or LN metastasis (Fig. 2). According to the adjusted multivariate Cox regression analysis, the type of surgical resection, sex, age, location, and cT category were not associated with RFS among patients with ypT0 to 1 disease (Table 3).
DISCUSSION Our current study findings indicate that PCRT can result in significantly higher TR of the primary tumor in rectal cancer patients and provide a greater influence on subsequent surgical
TABLE 1. Clinicopathological Characteristics of the Study Patients According to Their Clinical T Stage Category cT2 (n ¼ 100)
Variable Sex Male Female Age, y Tumor location (cm from AV) 5 >5 and 10 ycNþ Tumor regression grade TR No TR ypT ypT0–Tis ypT1 ypT2 ypT3 ypT4 ypNþ Surgery Local excision Radical resection
cT3 (n ¼ 465)
cT4 (n ¼ 85)
P 0.028
53 47 63
(53) (47) (38–79)
312 153 60
(67.1) (32.9) (35–78)
55 30 56
(64.7) (35.3) (32–69)
80 20 42
(80) (20) (42)
299 166 443
(64.3) (35.7) (95.3)
60 25 83
(70.6) (29.4) (97.6)
54 46
(54) (46)
82 383
(17.6) (82.4)
7 78
(8.2) (91.8)
60 19 21 – – 1
(60) (19) (21) – – (1.7)
87 18 145 217 1 113
(18.7) (3.9) (30.5) (46.7) (0.2) (24.3)
10 3 10 53 9 25
(11.9) (3.5) (11.8) (62.4) (10.6) (29.4)
42 58
(42) (58)
24 441
(5.2) (94.8)
0 85
(0) (100)
0.001 0.008
